Search

BackgroundBeing complex and highly heterogeneous with regard to the etiology and clinical manifestations of depression， neuroimaging studies make a breakthrough for exploring the biological subtypes of depression， while the current data-driven approach for the identification of subtyping depression using structural magnetic resonance imaging （MRI） data is insufficient.ObjectiveTo explore the biological subtypes of depression using diffusion tensor imaging （DTI） and machine learning methods.MethodsA total of 127 patients with depression who attended Beijing Anding Hospital from September 2017 to August 2021 and met the Diagnostic and Statistical Manual of Mental Disorders， fourth edition （DSM-IV） diagnostic criteria were included， and another 80 healthy individuals matched for gender and age were recruited through advertisements in surrounding communities during the same period. DTI findings， demographic characteristics and clinical data were collected from all participants. Tract-based spatial statistics （TBSS） and the Johns Hopkins University （JHU） white matter probability maps were used to extract fractional anisotropy （FA） values of white matter tracts. A semi-supervised machine learning technique was used to identify the subtypes， and the FA values for whole brain white matter of patients and controls were compared.ResultsPatients with depression were classified into two biological subtypes. FA values in multiple tracts including corpus callosum and corona radiata of subtype I patients were smaller than those of healthy controls （P0.05）， while subtype I patients scored lower on HAMD-17 than subtype II patients after 12 weeks of treatment （t=2.410， P

Satellite-derived bathymetry (SDB) techniques are increasingly valuable for deriving high-quality bathymetric maps of coral reefs. Investigating the performance of the related SDB algorithms in purely spaceborne active–passive fusion bathymetry contributes to formulating reliable bathymetric strategies, particularly for areas such as the Spratly Islands, where in situ observations are exceptionally scarce. In this study, we took Anda Reef as a case study and evaluated the performance of eight common SDB approaches by integrating Sentinel-2 images with Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2). The bathymetric maps were generated using two classical and six machine-learning algorithms, which were then validated with measured sonar data. The results illustrated that all models accurately estimated the depth of coral reefs in the 0–20 m range. The classical algorithms (Lyzenga and Stumpf) exhibited a mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) of less than 0.990 m, 1.386 m, and 11.173%, respectively. The machine learning algorithms generally outperformed the classical algorithms in accuracy and bathymetric detail, with a coefficient of determination (R2) ranging from 0.94 to 0.96 and an RMSE ranging from 1.034 m to 1.202 m. The multilayer perceptron (MLP) achieved the highest accuracy and consistency with an RMSE of as low as 1.034 m, followed by the k-nearest neighbor (KNN) (1.070 m). Our results provide a practical reference for selecting SDB algorithms to accurately obtain shallow water bathymetry in subsequent studies. [ABSTRACT FROM AUTHOR]

Lake Hulun is the fifth‐largest lake in China, playing a substantial role in maintaining the balance of the grassland ecosystem of the Mongolia Plateau, which is a crucial ecological barrier in North China. To better understand the changing characteristics of Lake Hulun and the driving mechanisms, it is necessary to investigate the water storage changes of Lake Hulun on extended timescales. The main objective of this study is to reconstruct the water storage time series of Lake Hulun over the past century. We employed a machine learning approach termed the extreme gradient boosting tree (XGBoost) to reconstruct the water storage changes over a one‐century timescale based on the generated bathymetry and satellite altimetry data and investigated the relationships with hydrological and climatic variables in long term. Results show that the water storage changes from 1961 to 2019 were featured by four fluctuation phases, with the highest water storage observed in 1991 (14.02 Gt) and the lowest point in 2012 (5.18 Gt). The century‐scale reconstruction result reveals that the water storage of Lake Hulun reached the highest point in the 1960s within the period of 1910–2019. The lowest stage occurred in the sub‐period of the 1930s–1940s, which was even lower than the alerted shrinkage stage in 2012. The predictive model results indicate the effective performance of the XGBoost model in reconstructing century‐scale water storage variations, with the mean absolute error of 0.68, normalized root mean square error of 0.11, Nash–Sutcliffe efficiency of 0.97, and correlation coefficient of 0.94. The annual fluctuations of water storage were mostly affected by precipitation, followed by vapor pressure, temperature, potential evapotranspiration, and wet day frequency. The dominating characteristics of different variables vary evidently with different sub‐periods. The atmospheric circulations of the Arctic Oscillation, El Nino Southern Oscillation, Pacific Decadal Oscillation, and North Atlantic Oscillation have tight associations with the water storage variations of Lake Hulun, which change with different study periods. Plain Language Sumamry: To better understand the changing characteristics of Lake Hulun and the driving mechanisms, it is necessary to investigate the water storage changes of Lake Hulun on extended timescales (e.g., century timescale). The main objective of this study is to reconstruct the water storage time series of Lake Hulun over the past one century. We employed a machine learning approach termed the extreme gradient boosting tree (XGBoost) to reconstruct the one‐century water storage changes and investigated the relationships with hydrological and climatic variables in long term. The century‐scale reconstruction result reveals that the water storage reached the highest point in the 1960s within the period 1910‐2019. The lowest stage occurred in the 1930s‐1940s, which was even lower than the alarted shrinkage stage in 2012. The annual fluctuations of water storage were mostly affected by precipitation, followed by vapor pressure, wet day frequency, potential evapotranspiration, and temperature. However, the dominating characteristics of different variables vary evidently with different sub‐periods. This study is expected to provide an efficient technical solution of reconstructing long‐term lake water storage records and to advance our scientific understanding of the characteristics of lake water balances in response to climate change and variability in the Mongolia Plateau. Key Points: The complete bathymetry map of Lake Hulun was constructed based multi‐source remote sensing dataA machine learning approach was employed to reconstruct the century‐scale water storage of Lake HulunThe potential links of one‐century variations of lake water storage with climatic variables and atmospheric circulations [ABSTRACT FROM AUTHOR]

Online multi-object tracking (MOT) is a challenging problem and has many important applications including intelligence surveillance, robot navigation, and autonomous driving. In existing MOT methods, individual object’s movements and inter-object relations are mostly modeled separately and relations between them are still manually tuned. In addition, inter-object relations are mostly modeled in a symmetric way, which we argue is not an optimal setting. To tackle those difficulties, in this paper, we propose a deep continuous conditional random field (DCCRF) for solving the online MOT problem in a track-by-detection framework. The DCCRF consists of unary and pairwise terms. The unary terms estimate tracked objects’ displacements across time based on visual appearance information. They are modeled as deep convolution neural networks, which are able to learn discriminative visual features for tracklet association. The asymmetric pairwise terms model inter-object relations in an asymmetric way, which encourages high-confidence tracklets to help correct errors of low-confidence tracklets and not to be affected by low-confidence ones much. The DCCRF is trained in an end-to-end manner for better adapting the influences of visual information as well as inter-object relations. Extensive experimental comparisons with state-of-the-arts as well as detailed component analysis of our proposed DCCRF on two public benchmarks demonstrate the effectiveness of our proposed MOT framework. [ABSTRACT FROM AUTHOR]

In this paper, an Interactively Recurrent Self-evolving Fuzzy Cerebellar Model Articulation Controller (IRSFCMAC) model is developed for solving classification problems. The proposed IRSFCMAC classifier consists of internal feedback and external loops, which are generated by the hypercube cell firing strength to itself and other hypercube cells. The learning process of the IRSFCMAC gets started with an empty hypercube base, and then all of hypercube cells are generated and learned online via structure and parameter learning, respectively. The structure learning algorithm is based on the degree measure to determine the number of hypercube cells. The parameter learning algorithm, based on the gradient descent method, adjusts the shapes of the membership functions and the corresponding fuzzy weights of the IRSFCMAC. Finally, the proposed IRSFCMAC model is tested by four benchmark classification problems. Experimental results show that the proposed IRSFCMAC model has superior performance than traditional FCMAC and other models. [ABSTRACT FROM AUTHOR]

Highlights: [•] A semi-supervised multi-graph hashing method is proposed for image search. [•] The different modalities are adaptively modulated by multi-graph learning approach. [•] Our hashing method integrates various modalities information with optimized weights. [ABSTRACT FROM AUTHOR]

Abstract: This study proposes a Rule-Based Symbiotic MOdified Differential Evolution (RSMODE) for Self-Organizing Neuro-Fuzzy Systems (SONFS). The RSMODE adopts a multi-subpopulation scheme that uses each individual represents a single fuzzy rule and each individual in each subpopulation evolves separately. The proposed RSMODE learning algorithm consists of structure learning and parameter learning for the SONFS model. The structure learning can determine whether or not to generate a new rule-based subpopulation which satisfies the fuzzy partition of input variables using the entropy measure. The parameter learning combines two strategies including a subpopulation symbiotic evolution and a modified differential evolution. The RSMODE can automatically generate initial subpopulation and each individual in each subpopulation evolves separately using a modified differential evolution. Finally, the proposed method is applied in various simulations. Results of this study demonstrate the effectiveness of the proposed RSMODE learning algorithm. [Copyright &y& Elsevier]

Abstract: This study presents a functional neural fuzzy network (FNFN) for classification applications. The proposed FNFN model adopts a functional neural network (FLNN) to the consequent part of the fuzzy rules. Orthogonal polynomials and linearly independent functions are used for a functional expansion of the FLNN. Thus, the consequent part of the proposed FNFN model is a nonlinear combination of input variables. The FNFN model can construct its structure and adapt its free parameters with online learning algorithms, which consist of structure learning algorithm and parameter learning algorithm. The structure learning algorithm is based on the entropy measure to determine the number of fuzzy rules. The parameter learning algorithm, based on the gradient descent method, can adjust the shapes of the membership functions and the corresponding weights of the FLNN. Finally, the FNFN model is applied to various simulations. The simulation results for the Iris, Wisconsin breast cancer, and wine classifications show that FNFN model has superior performance than other models for classification applications. [ABSTRACT FROM AUTHOR]

Abstract: Kernel-based nonlinear characteristic extraction and classification algorithms are popular new research directions in machine learning. In this paper, we propose an improved photometric stereo scheme based on improved kernel-independent component analysis method to reconstruct 3D human faces. Next, we fetch the information of 3D faces for facial face recognition. For reconstruction, we obtain the correct normal vector’s sequence to form the surface, and use a method for enforcing integrability to reconstruct 3D objects. We test our algorithm on a number of real images captured from the Yale Face Database B, and use three kinds of methods to fetch characteristic values. Those methods are called contour-based, circle-based, and feature-based methods. Then, a three-layer, feed-forward neural network trained by a back-propagation algorithm is used to realize a classifier. All the experimental results were compared to those of the existing human face reconstruction and recognition approaches tested on the same images. The experimental results demonstrate that the proposed improved kernel independent component analysis (IKICA) method is efficient in reconstruction and face recognition applications. [ABSTRACT FROM AUTHOR]

Abstract: The implementation of adaptive neural fuzzy networks (NFNs) using field programmable gate arrays (FPGA) is proposed in this study. Hardware implementation of NFNs with learning ability is very difficult. The backpropagation (BP) method in the learning algorithm is widely used in NFNs, making it difficult to implement NFNs in hardware because calculating the backpropagation error of all parameters in a system is very complex. However, we use the simultaneous perturbation method as a learning scheme for the NFN hardware implementation. In order to reduce the chip area, we utilize the traditional non-linear activation function to implement the Gaussian function. We can confirm the reasonableness of NFN performance through some examples. [Copyright &y& Elsevier]

In order to enhance the immune algorithm (IA) performance and find the optimal solution when dealing with difficult problems, we propose an efficient immune-based particle swarm optimization (IPSO) for use in TSK-type neuro-fuzzy networks for solving the identification and prediction problems. The proposed IPSO combines the immune algorithm (IA) and particle swarm optimization (PSO) to perform parameter learning. The IA uses the clonal selection principle, such that antibodies between others of high similar degree are affected, and these antibodies, after the process, will have higher quality, accelerating the search and increasing the global search capacity. The PSO algorithm has proved to be very effective for solving global optimization. It is not only a recently invented high-performance optimizer that is easy to understand and implement, but it also requires little computational bookkeeping and generally only a few lines of code. Hence, we employed the advantages of PSO to improve the mutation mechanism of immune algorithm. Experiments with synthetic and real data sets have performed in order to show the applicability of the proposed approach and also to compare with other methods in the literature. [ABSTRACT FROM AUTHOR]

This paper presents a self-organizing recurrent fuzzy cerebellar model articulation controller (RFCMAC) model for identifying a dynamic system. The recurrent network is embedded in the self-organizing RFCMAC by adding feedback connections with a receptive field cell to the RFCMAC, where the feedback units act as memory elements. A nonconstant differentiable Gaussian basis function is used to model the hypercube structure and the fuzzy weight. An online learning algorithm is proposed for the automatic construction of the proposed model during the learning procedure. The self-constructing input space partition is based on the degree measure to appropriately determine various distributions of the input training data. A gradient descent learning algorithm is used to adjust the free parameters. The advantages of the proposed RFCMAC model are summarized as (1) it requires much lower memory requirement than other models; (2) it selects the memory structure parameters automatically; and (3) it has better identification performance than other recurrent networks. © 2008 Wiley Periodicals, Inc. [ABSTRACT FROM AUTHOR]

In this study, an improved particle swarm optimization (IPSO)-based neural network controller (NNC) is proposed for solving a real unstable control problem. The proposed IPSO automatically determines an NNC structure by a hierarchical approach and optimizes the parameters of the NNC by chaos particle swarm optimization. The proposed NNC based on an IPSO learning algorithm is used for controlling a practical planetary train-type inverted pendulum system. Experimental results show that the robustness and effectiveness of the proposed NNC based on IPSO are superior to those of other methods. [ABSTRACT FROM AUTHOR]